The proposed research project is an inception cohort study of chemical incidents. The project is essentially an epidemiological study, which uses a mixture of qualitative and quantitative measures. Using a novel study design, the incident itself will be the unit of study rather than the exposed individuals. The project will be used to test the hypothesis that environmental incidents have a greater impact in affluent areas. In order to :do this a measure of 'perceived impact' of environmental incident on the exposed community will be developed and piloted, thereby allowing incidents to be designated as having either 'high' or 'low' impact. To assess the effect of the exposure of interest (socio-economic deprivation / affluence), standard socio-economic data will be obtained for the areas in which each incident occurs. There are likely to be many important effect modifiers and confounders in determining outcome after such incidents. Data will be collected on the toxicity of the chemical incident, but also on factors affecting risk perception (e.g., presence of an odour and media coverage) and other social factors. Multi level modeling will be used to assess the relative importance of these diverse factors. [unreadable] [unreadable]